made some changes for faster processing

app.py

@@ -47,7 +47,7 @@ def process_text(text):
     return emotion_dict, predicted_emotion
 
 def preprocess_frame(frame):
-    frame = cv2.resize(frame, (224, 224))
+    frame = cv2.resize(frame, (112, 112))
     pixel_values = caption_processor(images=frame, return_tensors="pt").pixel_values
     return pixel_values
 
@@ -95,7 +95,7 @@ def analyze_video(video=None, video_url=None):
         emotion_dict_text, predicted_emotion_text = process_text(transcript)
 
         # Frame-wise emotion detection from the video
-        n_frame_interval = 60
+        n_frame_interval = 120
         emotion_vectors_video = []
 
         video_capture = cv2.VideoCapture(video_path)